1. Field of the Invention
This invention relates to a recording/reproducing apparatus which records and reproduces digital signals, and more particularly to the digital data encoding/decoding apparatus for optical disk apparatus which carries out mark edge recording by pit width modulation.
2. Related Art of the Invention
In recent years, the optical disk apparatus has attracted keen public attention as a digital information recording/reproducing apparatus with a large storage capacity and the ability to interchange media. In the optical disk apparatus, the digital data is encoded for recording in such a manner that it fits in the recording/reproducing channel characteristics determined by the optical head and optical recording media, and reproduced signals are data-detected to make binary-coded signals, from which decoding is carried out to obtain original digital data. Based on this, encoding and decoding techniques to enable efficient digital recording and reproducing have been put into practical use as various digital data encoding/decoding apparatus.
In the general procedure to record and reproduce digital data, first of all, the code stream obtained by encoding the data with modulation having a proper rule is recorded. When reproducing, the clock signals which are clock components of the code stream are retrieved from the reproduced signals by utilizing the properties of the code stream imparted by the above-mentioned modulation. Based on the retrieved clock signals obtained, the recorded code stream is separated and the original digital data is obtained by decoding, which is the operation reverse to encoding. As one example, in the standard format of a 130-mm-diameter magneto-optic disk data file apparatus, a (2, 7) code is used as the encoding system as found in the International Standard (ISO/IEC DIS 10089). Table 1 shows the conversion rule of the (2, 7) code.
The (2, 7) code is an encoding system which converts 1-bit digital data to 2-bit codes, and is so called because of its characteristic that a "1" is separated by a minimum of two "0's" and a maximum of seven "0's" in the code stream after encoding an original digital data stream. This encoding rule is called (d, k) conversion rule because a "1" is separated by a minimum of d "0's" and a maximum of k "0's" in the code stream after encoding. Consequently, in the case of the (2, 7) code, a "1" exists intermittently at the clock frequency from 3 to 8 in the code stream after encoding, and with this point as a premise, the code frequency which is a clock component is obtained to make a reproduced clock signal, rendering itself capable for detecting the above-mentioned data.
In general, in this type of encoding/decoding, decoding timing to the bits of the code stream and the data stream is to be properly provided when the original data stream is decoded from the reproduced code stream. Otherwise, such decoding timing failure does not keep the rule of the encoding system and will result in errors. In the case of Table 1, the 1-bit data must be made to correctly correspond with the 2-bit code. For this purpose, a specific code pattern called SYNC or RESYNC BYTE is inserted in the code stream after encoding to achieve bit synchronization at the time of decoding. In the above-mentioned international standard, for the RESYNC BYTE, EQU {0 0 1 0 0 0 0 0 0 0 1 0 0 1 0 0 }
is periodically inserted, and in the event of decoding operation, first, this RESYNC BYTE is detected; then, based on this, the above-mentioned bit synchronization is achieved for decoding. This RESYNC BYTE is a pattern which cannot exist in any code stream after encoding for any data stream in the rule specified in Table 1, and the RESYNC BYTE will never be detected by mistake with other code data.
This kind of method enables this bit synchronization by periodically arranging bit patterns, which can easily be identified, as a RESYNC BYTE to achieve bit synchronization in the encoded code streams as described in, for example, OPTICAL DATA FORMAT EMPLOYING RESYNCHRONIZABLE DATA SECTORS, U.S. Pat. No. 4,791,622 by D. W. Clay et al. and SYNC ENCODING SYSTEM FOR DATA SECTORS WRITTEN ON A STORAGE MEDIUM, U.S. Pat. No. 4,797,167 by M. J. O'Keeffe et al. This bit synchronization pattern is called a SYNC BYTE when it is used at the data head, and a RESYNC BYTE when used at the intermediate position. The SYNC BYTE decides the bit synchronization at the start of decoding, while the RESYNC BYTE periodically corrects deviation of a decoding bit to prevent propagation of decoding error after any defect occurs when clock reproduction failure occurs in the middle of data reproduction, and both frequently have the same patterns.
In the meantime, in the present optical disks, a large number of developments have been undertaken to increase the capacity to store more and more information, and in order to avoid complication of data controls associated with the increased storage capacity, the unit of the recording data amount must also be increased. When the recording data unit is increased, there will be more possibility to cause failure to reproduce clocks during data reproduction due to the drop-out of reproduced signals arising from a defect of the media, and the system reliability will be lowered. The importance of the RESYNC BYTE has been further increased to suppress continuous occurrence of decoding errors in order to prevent disability in decoding all of the data after the clock reproduction failure occurs.
In the conventional optical disk apparatus, as a method to record and reproduce data, mark position recording (MPR) in which the recording mark position is used for information recording is carried out; this is called pit position modulation (PPM) because of the recording pit, another name of the recording mark, and has characteristics to correctly record and reproduce the data even when there is a variation in pit size. However, in order to further increase the recording density, the pit width modulation (PWM) which carries out mark edge recording (MER), in which the position and length of the recording mark are used for recording of information, has begun to be put into practical use.
For the encoding rule in the PPM recording system, the (2, 7) code has the superior capabilities, and the PPM recording system using this (2, 7) code has been adopted in the above-mentioned international standard, but as part of further improving the recording density, in the PWM recording, investigation has been made on systems such as a (1, 7) code. In this mark edge recording, NRZI code is performed after a (2, 7) or (1, 7) code. The positional relationship between the mark formed in correspondence with the part in which "1" of the code stream obtained as above continues and the space other than this formed by "0" is used for recording information.
The mark is formed by applying the comparatively high optical output to the medium and locally raising medium temperature, and has a problem that the positional relationship of mark edge deviates when the optical output deviates from the optimum value. That is, when the optical output is greater than the optimum value, the recorded mark generally becomes larger, and, on the contrary, when it is smaller, it becomes smaller, causing the power margin, the set margin for optical output in recording, to become smaller. In this way, the positional relationship between the mark initiation end and the finish end deviates from the optimum bit intervals of code data, resulting in higher possibility to cause decoding errors in achieving synchronization of data from reproduced signals for decoding.
When the (2, 7) code is used as a technique to solve this kind of problem, for example, as is found in DATA RECORDING/REPRODUCING DEVICE, U. S. Pat. No. 5,229,986 by Mizokami et. al., specially designing the data detection method during reproduction has enabled the development of a technique to improve the detection allowance for mark formation. In this method, synthesis of both takes place after the mark initiation end and the finish end are independently binary-coded and clock-reproduced, and both are designed to achieve bit-synchronization independently. Consequently, the data detection method is limited and it is not generally applicable.